The death of a loved one is one of the most distressing life events that greatly affects physical, social and psychological well-being. Numerous studies, in fact, have shown that bereaved individuals have higher rates of disability and medication use than their non bereaved counterparts, and are themselves at heightened risk of death (Zisook and Shear, 2009). Despite most individuals adequately adjust to bereavement, there are many ways in which people cope with the loss of a loved one. Even if there is no right way of coping with death, everyone develops a personal experience depending on his/her own personality and on the relationship with the person who died (Prigerson et al., 2009).
There is agreement in the literature that bereavement is associated to an excess morbidity and mortality (Jacobs, 1993; Stroebe and Stroebe, 1993; Latham et al., 2004; Onrust et al., 2007) and may trigger an increased risk for major depression and anxiety disorders (Thompson et al., 1984; Bruce et al., 1990; Clayton, 1990; Jacobs et al., 1990; Zisook et al., 1993; Schaefer et al., 1995; Parkes, 1996; Lichtenstein et al., 1998; Murphy et al., 1999; Shahar et al., 2001; Schultz et al., 2001; Li J et al., 2003; Bradbeer et al., 2003; Christakis et al., 2006; Onrust et al., 2006; Stroebe et al., 2007).
Many studies have been recently focused on the distinction between normal and complicated grief (CG) (Chiambretto, 2008; Prigerson et al., 2009; Zisook and Shear, 2009), the latter also called traumatic grief (Prigerson et al., 1999; Jacobs et al., 2000; Prigerson et al., 2001) and, more recently, prolonged grief disorder (Chiambretto, 2008; Prigerson et al., 2009). Grief is an unavoidable and normal reaction to a loss but if the normal grief reactions increase, becoming increasingly debilitating, with intrusive thoughts, sad images of the deceased person, with painful yearning inducing a significant impairment in work and social functioning, it can be defined as CG (Prigerson et al., 2009). CG may occur in about 9 to 20% of subjects (Zisook and Shear, 2009; Kersting and Kroker, 2010) facing bereavement and research efforts have been recently devoted to clearly identify its clinical characteristics.
CG is characterised by specific patterns of symptoms that can persist for years and resolve slowly such as symptoms of separation distress from the deceased (e.g., longing, yearning and searching behaviours for the deceased, excessive loneliness and frequent intense pangs of grief and sadness related to the loss) or of traumatic distress (e.g., feelings of disbelief, daydreams about what was lost, a need to remember, fragmented sense of security and trust, intense and prolonged preoccupation with thoughts of the deceased, avoidance and failure to adapt).
CG symptomatology has repeatedly been shown to be different from symptoms of other DSM-IV disorders (Prigerson et al., 1995; Prigerson et al., 2009), consisting of core symptoms of separation and traumatic distress, and CG has been discussed as a stress related disorder distinct from major depression (MDD), adjustment disorder (AD), acute stress disorder (ASD) and PTSD (Horowitz et al., 1993, 1997; Prigerson et al., 1999; Jacobs et al., 2000; Stroebe et al., 2000; Prigerson and Jacobs, 2001; Lichtenthal et al., 2004).
Distinguishing CG and MDD may be difficult because of the frequent co-occurrence of the two conditions (Prigerson et al., 1995, 1996; Lichtenthal et al., 2004; Maytal et al., 2007; Brent et al., 2009; Kersting et al., 2009) that may co-exist in percentages ranging between 52 and 70% (Melhem et al., 2001; Simon et al., 2007). Assessment of CG and depressive symptoms has revealed that although they are frequently comorbid, they can occur independently (Enright and marwit, 2002, Horowitz et al., 1993; Prigerson et al., 1997), with a different course (Pasternak et al., 1991, 1993; Prigerson et al., 1996; Prigerson et al., 1995; Zisook and Devault, 1983) and a different response to treatment (Prigerson et al., 1995). In studies of clinical populations, CG has been associated with high rates of suicidal ideation, an history of suicide attempts and indirect suicidal behaviours, not explained by co-occurring major depression, and with elevated rates of lifetime suicide attempts (Prigerson et al., 1999; Szanto et al., 2006).
Thus, the primary aim of the present thesis was to explore the prevalence and impact of lifetime mood spectrum symptoms, as assessed by a validated instrument such as the Mood Spectrum Self Report (MOODS-SR) lifetime version, in a sample of patients with CG. In particular, we explored the impact of mood spectrum symptoms, including depressive, manic, and rhythmicity and vegetative function symptoms, on suicidality (investigated as suicidal ideation and attempts) in these patients.
Secondary aim of the present thesis was to investigate the prevalence and impact of mood spectrum symptoms and adult separation anxiety disorder symptoms in patients with CG with respect to patients with PTSD or a comorbidity between the two diagnoses.
The whole sample included 116 out- and in-patients with a diagnosis of CG and/or PTSD, recruited at the Department of Psychiatry, Neurobiology, Pharmacology, and Biotechnology of the University of Pisa, Italy, from the clinics and hospital wards involved. All patients were assessed by means of the Structured Clinical Interview for DSM-IV Axis-I disorders Patient version (SCID-I/P) for Axis I disorders diagnoses (First et al., 1995); the MOODS-SR lifetime version, a questionnaire for the assessment of mood spectrum symptomatology (Dell’Osso et al., 2002), the Inventory of Complicated Grief (ICG), a self report instrument to assess maladaptive symptoms (Prigerson et al., 1995), the Adult Separation Anxiety Disorder questionnaire (ASA-27), an instrument used to check adult separation anxiety (Manicavasagar et al., 2003), the Impact of Event Scale (IES) (Horowitz et al., 1979) and the Work and Social Adjustment Scale (WSAS) (Mundt et al., 2002)
The results of the present thesis, in accordance to literature data (Prigerson et al., 1997; Mehlem et al., 2001; Piper et al., 2001) show comorbid mood disorders in the majority of our CG patients. However, it is noteworthy that the comorbidity with anxiety disorders was quite frequent: to our knowledge, this is the first study highlighting the presence of different anxiety disorders in CG, since previously the comorbidity with MD or PTSD was underlined (Prigerson et al., 1997; Zisook and Shear, 2009).
In our sample, high levels of suicidality were observed in patients with CG: in fact, more than a half of them reported suicidal ideations and almost one fourth suicide attempts. This finding is in agreement with those of previous studies indicating that bereaved subjects who meet criteria for CG had a heightened risk for both suicidal ideation and attempts (Stroebe and Stroebe, 1993; Rozenzweig et al., 1997; Latham and Prigerson, 2004; Szanto et al., 2006; Prigerson et al., 2009). Subthreshold depressive and rhythmicity/vegetative functions items of the MOODS-SR were significantly associated with increased suicidal ideation and attempts, while subthreshold manic items to suicidal ideation only. Relationships were confirmed after controlling for axis-I disorders comorbidity.
When comparing our sample of patients with CG with a sample of subjects with PTSD we divided the former group on the bases of the presence of a comorbid diagnosis of PTSD. Thus, the sample of CG patients resulted to be composed by two subgroups: one including 28 subjects, who did not present any comorbidity for PTSD, and another including 22 subjects who presented both disorders.
We observed significant differences amongst the three diagnostic groups in lifetime comorbidity for mood disorders. Almost two thirds of patients with CG or both disorders reported a diagnosis of major depression with respect to about half of those with PTSD only. Patients with both disorders presented the highest rate of bipolar disorder comorbidity. Interestingly, exploring the sample through the MOODS-SR, the highest scores on the lifetime depressive component were reported by patients with both diagnoses, while the lifetime manic component was significantly higher in patients with PTSD alone and CG+PTSD than in patients with CG alone. Even the scores of the scales assessing post-traumatic stress symptomatology or adult separation anxiety symptoms were significantly higher in patients with both disorders with respect to patients with either CG or PTSD alone.
Our results show different levels of lifetime depressive and manic spectrum symptoms on patients with CG with respect to those with PTSD or both disorders corroborating the evidence of CG as a distinct condition. Mood spectrum symptoms also seem to play a more relevant role on the risk for developing CG and or PTSD with respect to separation anxiety disorder that show high rates only when the two conditions co-occur.
From such a perspective, our data highlight the need to comprehensively explore lifetime subthreshold mood symptoms in CG patients in order to promptly identify those who may be more prone to suicidal ideation and/or attempts. Furthermore, the results of the present thesis highlight the need to comprehensively explore lifetime subthreshold mood symptoms in patients who experience a loss or a traumatic event, in order to identify those subjects who may be more prone to develop enabling mental disorders such as CG or PTSD.